VOL. ?, NO.

3 WATER RESOURCES RESEARCH 3UNE 1971

Alternative
Oxygenation
Possibilities LargePollutedRivers

WILLIAM WHIPPLE• JR.• AND SHAW L. YU

Water ResourcesResearchInstitute, Rutgers University, New Brunswick,New Jersey 08903

Abstract. The idea of induced aeration of polluted rivers is gaining acceptance,prototype

equipment having been installed on the Miami River in Ohio and on the R.uhr River in
Germany. Previous anglysisand reports have indicated the general feasibility of aeration
on small rivers. This paper describestests of aeration equipment on a major river, the Dela-
ware Estuary, and discussestransfer efficiency,suita.bleaerator system design, and cost com-
parisons. Navigational considerationsand deeper, more turbulent water cause design re-
quirements for large rivers to differ considerably from the requirements for small rivers.
Reinforced aeratorsat the surfaceappear favorable for uncrowdedriver areas,but air diffusers
on the bottom appear more practicable and comparably economical for port areas. Diffusers
½gnbe placedonly on channelmarginsand anchorageareas,but dispersionstudiesindicate that
the oxygenated water will still reach the center of the river within a reasonable distance. The
cost of adding one unit of dissolvedoxygen by aeration devices appearsto be just about one-
fourth that of adding it by officially approved waste treatment only. There are possibilitiesof
lowering costsfurther by using oxygen diffusers.Institutional barriers that make it difficult to
plan and develop pollution control alternatives suchas fiver aeratorsare discussed.

The idea of artificial or induced aeration of regionalwater quality control agency,the Dela-
polluted rivers has been developing fairly ware River Basin Commission, and by the sub-
rapidly in recent years. Analysts have com- missionin 196.6of a special report highlighted
pared the possibilitiesof induced aeration to by the commissionerof the former Federal
the standard approachesof advanced effluent Water Pollution Control Administration [1966].
treatment and low flow augmentation [Kneese More recent studies are now available [Wright
and Bower, 1968; Davis, 1968]. Tests and cost and Porges, 1970']. About 40 miles of the tidal
studies of prototype equipment under scien- section of the Delaware suffers at times from
tifically controlledconditionshave demonstrated insufficientdissolvedo.xygen(DO) (Figure 1).
the applicability of instream aeration to situa-The oxygendeficiencyinterfereswith important
tions on small polluted rivers, especiallythe runs of anadromousfish, besidesimpairing the
PassaicRiver [Whipple et al., 1969; Hunter quality of the river for recreational and living
and Whipple,1970; Imho#, 1968; Tarassovet purposes.The adopted water quality standard
al., 1969;DavidsonandBradshaw,1970; Whip- will require bringing the minimum daily aver-
ple et al., 1970]. Meanwhile the Ruhrverband age DO levelsup to 3.5 rag/1 at a costof about
and more recentlythe Miami Conservancy Dis- $40 million annually. Annual cost estimatesof
trict have installedaeratorsand are beginning $46 million and $34.4 million correspondto two
to obtain operatingresults.This paper describes given objectives between which the adopted
tests of prototype aeration equipment on a water quality standardslie [Wright and Porges,
major river, the Delaware Estuary, and dis- 1970]. Comparisonof these two objectivesin-
cussesmethods of determining transfer effici- dicates that adding the last 1.0 mg/1 of DO
ency, suitable aerator systemdesign,and rela- to the critical section of the river will add about
tive costs. $9.3 million to systemcostsannually.Sincesuch
great costs are projected for the officially
TI-IE DELA•VARB ESTUARY PROBLEM
adopted plan and sincethat plan is so clearly
The lower Delaware River, known as the defined,the Delaware obviouslyconstitutesan
Delaware Estuary, has been recognizedas a ideal area for investigatingpossiblealternatives
majo,r water quality problemarea by the crea- for achieving given water quality objectives.
tion of the nation's only fully empowered The Delaware River is more than •-mile
566
 Artificial Aeration o• Rivers 567

wide in the critical section,is tidal and heavilytion downstreamthrough which all or most of
navigated,and has a channelmore than 40 feet the aerated water may be assumedto pass, as
deep. Design and operating conditionsfor such determined by observing the flow patterns
a river must be quite differentfrom those for through fluorescentdye tracing. Differences in
a small river like the Passaic or from those for the total DO passingthrough the control sec-
aeration lagoonsand treatment plants, in which tion before and after the aerator is put into
aeration equipmentis commonlyused.This fact operation were attributed to the aerator, subject
was dramatically emphasizedwhen, on the day to verification from an upstream section that
after its installation, the frame of the surface the DO from upstream had not changed.This
aerator was buckled by waves in a sudden approach may be expressedas follows.
squall. The aerator had to be removed,hastily Considera small area A At, within which the
redesigned,and rebuilt. The main test pro- DO concentrationsmay be considereduniform.
gram was carried out during the summerof 1969' These concentrationsare representedby C•. If
in the area adjacent to the pier of the Camden the velocity of flow for the small area AA• is
sewage treatment plant. Figure 2 shows the V•, then the massrate of oxygenpassingthro.ugh
1969 test site, with the surface aerator operat- the area is C•V• A At. The total mass rate for
ing. In June 1970 further tests of air diffusers the control section M is then
were conductedfrom a pier of the Philadelphia
Port Authority to extend the range of tests of
diffuser equipment to deeper water. Let C• be the DO concentration before and
C/ the DO concentration after the aerator is
AERATOR OXYGEN TRANSFER PROCEDURES
put into operation. The oxygen uptake rate
The aerators tested in the Delaware River due to the aerator U can be computedby
are shown in operation in Figures 3 and 4.
Methods of evaluatingthe oxygentransfer (up- = (c,' - c,)
take) rate for such aeratorson small rivers are under the assumption that there are no sig-
based on differencesin the total oxygencarried nificant velocity changesat the control section
by the stream crosssectionsabove and below before o.rafter the aerator is put into operation.
the aerator at any given time [Whipple et al., The oxygen transfer rate for the aerator
1969]. A different method is required for large under test conditions R, can then be determined
rivers. This method is based on a control sec- by

mg/l
DO [ [
Zone
2 --i• __
Zone
3 • [ Zone
____ 4 _•__ Zone

I I
'•
x k
'[, stream 'l
eriteri•
', '/
I k k Mi•mum
M•mu •fiy • • k [ average

Fig. 1. Dissolved oxygen criteria and recent conditions [Craine, 1969].

568 WHIPPLE •,ND YU

•, = v/•' (3) small amount o.f oxygen transfer during the

bursting phase of bubble aeration in conven-
where P is the power consumption of the tional aeration tanks, other investigators
aerator in shaft horsepower. [Barnhart, 1969] have observed a substantial
The mechanismsof oxygen transfer are dif- amount of oxygen transfer at the surface in
ferent for diffuser and mechanical aerators. For
highly turbulent aeration tanks. Therefore for
diffuser aerators the total rate of oxygen trans- diffuserswith large aperturesthe total rate of
fer is made up of three components:bubble oxygen transfer N may be written as the sum
formation,bubble a•eent, and bubble breakage of two individual rates, i.e.., the rate during
at tile water surface [Bewtra and Nicholas, bubble ascentNa, and the rate during bubble
1964]. For porous diffuserswith small capillary burstingat the surfaceN,:
openingsand small airflow rates, oxygentrans-
fer during bubble formation is appreciable.
However, for diffuserswtih large apertures,like The oxygen transfer rate during bubble as-
the one testedin this study, bubblesare formed cent.N• dependson such factors.as the depth
while they are rising in the water, and oxygen of diffuser submergence,the velocity of the air
transfer during bubble formation is negligible. bubbles, the diameter of the air bubbles, and
Aeration during bubble bursting at the water so.forth. On the other hand, N, has been ex-
surface is related mainly to turbulence con- pressedas an exponentialfunction of the water
ditions. Although Carver [196.9'] observed a velocity at the surface [Barnhart, 19'69.].It is

Fig. 2. Aerator test site at Camden, New Jersey.

 Ar&iy•cialAera•io• o] Rivers 569

Fig. 3. Reconstructed surface aerator on the I)elaware River.

therefore expected that for • diffuser the ef- rioned phasesof oxygen transfer, i.e., spray
ficiency o,f oxygen transfer would be propor- aerationand turbulent mixing.
tional to the depth of submergence, the size.of In this study aera.torswere operated and
the bubbles, the airflow ra.te, a.nd the surfa.ce data were collectedonly on outgoing tides to
turbulence [Eckenfelder, 19'5.9; Eckenfelder avoid the effects of tidal current a.nd channel
and Ford, 1968]. asymmetry.At ea.chcrosssectionboth upstream
For mechanical aerators oxygen transfer to and downstrea.m from the aerator velocities at
the Wa.teroccurs.
both in the sprayedwater and va.rio,uspo,sitionsa.nddepths were obtainedby
in the turbulent mixing zone [Eckenfelder and current meter, and correspondingDO readings
Ford, 1968]. Investigation of the process of were measuredby oxygenprobescalibratedby
spray aeration shows that oxygen transfer is frequent Winkler tests. The a,verageDO con-
related to' the size of the water droplets centra,tionfor a sectionwas obtainedby weight-
[Carver, 1969']. For surface entrainment aera- ing the various point readingsaccordingto. the
tors the efficiencyo,f .oxygentransfer generally velocities.,and the DO increment.was obtained
decreaseswith increasingaeration basin volume by equation 2'.
but increaseswith increasingwater depth for For the deepwater air diffuser tests under-
basins o,f co.nsta.ntdiameter [Garland, 1969]. taken at the Philadelphiasite in June 1970, the
To date, however, no information is available proceduresdescribedabove were impracticable
concerningindependentevaluatio.nof the rela- beca.use of the much smaller aerator ma•fold
tive importance of each of the two. aforemen- used. Therefore the oxygen absorption was
570 WHIPPLE AND YU

determinedby taking air samplesfrom diffuser in which

bubblesrising to the surfaceand then analyzing
their oxygen content. The analysis was con- (C•)•.o, saturationDO concentrationat 20øC;
ducted by gas chromatograph,a gas sampling (C•)•, saturation DO concentrationat test
valve being used. These results,when checked water temperatureT;
with measurementsfor atmospheric. air samples, P, pressurein inchesof mercury;
are consistent with the results obtained for the fi, specificsolubility of oxygen;
C•, DO concentrationat the aerator;
lesserdepths by other methods; thus the pro-
cedure is shown to be reliable. TF, temperaturecorrectionfactor, equalto
(1.025)•-•-0;
Oxygen transfer rates Rt were reduced to
a, specificoxygentransferrate;
those under standard conditionsR8 by the fol-
F, conversionfactor.
lowing relationship,so that comparisonscould
be made with results elsewhere: The saturation oxygen concentrationC• was
computedby the SanitaryEngineeringDivision,
(5)
American Society of Civil Engineersequation
[Whipple et al., 19'69].For the diffuseraerator,

F--i(C,)•(2•P.
92)(fi).
CmI
(TF)(o•
) C8was correctedfor pressure,whichin turn was
evaluated at the midpoint of the depth of sub-

.:.

........
::•i•!!i•::•::i
........
:....::?•::.. :•.::•::::•.:.

'" ' ...........

'•:::•:•q•'::•'
""'::•-•'•'•.-:-"•i•:•
.......
'.......
:•..i:::

....

Fig. 4. Diffuser aerator upwelling.

 Artificial Aeration o[ Rivers 571

Table 1. Gas SamplesTaken July 2, 1970, in Philadelphia

CorrectedFlow % O• in % O•
Sample No. Header Depth Rate, scfm SpentAir Absorbed
1 12'3" 93.0 20.8 1
2 12'3" 93.0 19.9 6.5
3 12'3" 44.0 20.06 5.6
4 12'3" 44.0 20.4 3.6
5 12'3" 16.3 19.9 6.5
6 12'3" 16.4 19.91 6.5
7 38'3" 19.7 20.7 1.8
8 38'3" 19.7 19.07 11.3
9
10 38'3" 53.7 19.6 8.2
11 38'3" 53.5 18.9 -----12.3
12 38'3" 99.4 19.3 10.0
13 38'3" 99.4 19.7 7.7
14 25' 17.9 20.1 5.3
15 25' 17.9 19.8 7.1
16 25' 49.6 19.6 8.2
17 25' 49.6 19.7 7.7
18 25' 49.6 19.7 7.7
19 25' 94.1 20.3 4.2
20 25' 92.9 19.4 6.0

Air intake samples

A• 21.1
A• 21.0
A3 20.9
A4 20.83

Air-filled sample bulbs

N•* nO. 7
N• 0.0

* Purged with Na 15 seconds.

t Purged with Na 30 seconds.

mergence. According to laboratory tests car- ever,the smallnumberof testsand the difficult
ried out by J. V. Hunter but not yet reported, conditionsfor measuringresultsreducethe con-
both a and • valueswere found to be 1.0. fidence that can be accorded to these results.
Many more data were obtainedwith the dif-
OXYGEN TRANSFER I•SULTS
fuser aerator than were obtained with the
Owing to a late start of the tests and the con- mechanicalaerator. To supplementresultspre-
sequencesof storm damage,only four complete viously obtainedon the PassaicRiver at depths
sets of results were obtained for the mechanical of about 8 feet, tests were performed at the
aerator, which is only a fraction of the data Camden site at depths varying between 11.0
previoudy obtained in shallower water. The and 16.9 feet becauseof the tidal range. At the
computedoxygentransfer rate at standard con- Philadelphiasite tests o.f a similar but shorter
ditions varied from 1.29 to 4.50 lb/hp hr, the diffuser were made at 12.3, 25, and 38.3 feet.
average being 3.06 lb/hp hr as comparedwith A summary of the results is given in Table 1.
an average of 2.12 lb/lap hr obtained on the A study of these data showsthem to be quite
PassMe River. Some increase in the transfer consistentdespitethe differentperiodsof time
rate might be expectedbecauseof the greater and methods of measurement involved. The
volume of water affected in a deep river and proportion of oxygen absorbed obviously in-
the probability of a reducedtendency for re- creaseswith depth,but the increaseis muchless
circulation of previously aerated water. How- than proportionate,particularly at the greater
572 •I-IIPPLE AND YU

depths.Thus the oxygentransferrate in pounds In the diffusionresultsreported here no cor-

per horsepower-hourshowsa considerablede- relation was found between the percent of
crease.at greater depths. The o.xygentransfer oxygen absorption and the rate of airflow, al-
rat©--{Figure5) is indicatedto be 1.29 lb/hp hr though such a relationship has been reported
at 7.2 feet, 1.36lb/hp hr at 13.2feet, 0.93 lb/hp previously for other diffusers [Eckenfelder,
hr at 25.0 feet, and 0.68 lb/hp hr at 38.3 feet. 19'59; Bewtra and Nicholas, 1964]. The lack
Figure 5 illustrates this variation of transfer of correlation may be due to the small range
rate with depth of submergence. o.f flow ra.tesused. It may also be that within
To comparethe resultswith thoseof other the rangestested,altho.ughthe rate of absorp-
investigators,
the oxygenuptakes obtainedin tion increaseswith the rate of flow, the velocity
this study were plotted in Figure 6 in terms of rising bubbles also increases,and a shorter
of percentof oxygenabsorptionand depth of elapsed time for absorption results.
diffusersubmergence. Results for saran tubes
SYSTEMS DESIGN PROBLEMS
and spargertype diffusersoperatingin aeration
tanks for airflow rates rangingbetween10 and The aeration equipment tested on the Dela-
15 scfm (standard cubic feet per minute) per ware is readily available commercially, since
unit [Bewtra and Nicholas, 1964] appear to aerators are widely used in activated sludge
showa linear relationshipbetweenpercentof processesand aeration lagoons. As previously
absorption
anddepthof submergence
whenthe mentioned the mechanical aerators used were
resultsare plottedon log-logpaper.The oxygen st.ructurally inadequateto resistwave action on
transfer rates indicatedare considerablyhigher large rivers, but redesignfor this purposewould
than thoseof the presentstudy,owing,in part not be difficult or expensive. The Ruhrver-
at least, to the fact that bubblesrise at a band [Imhol•, 1968] has used mechanicalaera-
slower rate in tanks than in laterally unre- tors successfullyon Lake Baldeney, an im-
stricted bodiesof water. However, diffuserson poundedsectionof the Ruhr River. In that case
the Ruhr River alsoare reportedto have shown two aerators mounted at opposite ends of an
much higher efficiencies than those on the anchored raft are driven by a diesel engine in
Delaware River [Imhoff, 1968]. In the Ruhr the center (Figure 7). The Ruhrverband aera-
much"finer bubbleswere used, the orificesin tor impellers,similar to thoseused by Rutgers
Dr. Imhoff'splastictubesbeing0.5 and 0.7 mm researchers,are large and slow moving and
in diameteras compared with 5/32 inch (about handle a great deal of water at relatively low
4.0 mm) for thoseusedin the Delaware. velocity. The Miami ConservancyDistrict has

2.0

1968 PASSAIC DATA

1969 DELAWARE DATA

1970DELAWARE
DATA
1.5

0.5

10 2o 3o 0
Depthof Submergence,ft

Fig. 5. Oxygentransferrates versusdiffusersubmergence.

 Artificial Aerationo• Rivers 573

ß Imhoff

2O

--•
-
•
_•Rutgers,
Saran
tubes}
Bewtra
Sparget s &
Nicholas
Passaic
-

-
(D Rutgers, Delaware
--
(gas samples)
-• Rutgers, Delaware
(mass balance)
o

ß
• lO
o

, ,

4 5 6 7 8 9 10 20 30 4o

Diffuser Submergence, Ft

Fig. 6. Oxygenabsorption
versusdiffusersubmergence.

recentlyinstalleda groupof four electricdrive wouldneedto be far enoughout into the water
surface aerato.rs on the Miami River near that natural currents and tides would disperse
Dayton, Ohio.Theseaeratorshave a relatively the water throughoutthe affectedcrosssec-
small high-speedimpeller mountedin the cen- tion of the river. This requirementis,essential
ter of a doughnut-shaped plastic float. They because research on anadromous fish of the
appearto have efficienciesroughlycomparable Delaware indicates that these fish do not per-
to those covered in this report. They are ceivelow oxygenwatermasses, andwhenthey
moored to cablesstrung acrossthe river, up- beginto suffocate
they reactrandomly, ,being
stream and downstream (Figure 8). as apt to swimdeeperinto the bad area as to
On a large highly developedriver like the withdraw [Dorfman and Westman,1970].
Delaware,'the requirements of navigationare Aerators create strong local turbulencein
controlling.The most practicablemooringar- the river, and dispersion
studieson the-'•Dela-
rangement for a surfaceaeratorappearsto be ware basedon fluorescentdye tracingof,aerator
'threepile clusters,
eachconnected to the aera- effects have shown that the' aerated water
tor by cables(Figures9 and 10). Electricserv- would extend out from each side to. the center
ice wouldbe providedby submarine cable.It of the river within 5.7 miles of the aeration
appears that the mostsatisfactorygeneralplan site (GeorgeMattingly of PrincetonUniversity
is to spacethe aerationsitesof oneor several and John B. McCall,P. W. Anderson, and John
unitsalong'eachbank'of the river in' areas Murphyof the U.S. Geological Survey,unpub-
wherenavigationSusage will allow.Suchunits lisheddata, 1969).
574 WHIPPLE AND ¾U

The aeration system outlined above, based

on fixed aeratorsites,wouldnot be suitablefor
a stagnantriver or a waterway with little cur-
rent like the Houston ship canal. The Black
Warrior River at Tuscaloosais a special case;
there normal flow ceasesfrom Friday afternoon
to Monday morningbecauseof peak operations
of a large hydroelectricplant on the dam up- \
I
stream. For such cases if induced aeration were I
necessary,moving aeratorscould be used.See
Figure 11 for a possibledesignof sucha unit. ii
Diffuser aerators as well as surface aerators
are subjectto constraintsimposedby naviga- • II
II
tional requirements.Whereasfixed surfaceaera-
tors cannotbe usedat all in port areasand other •11
intensivelyused portionsof the river, diffuser III
aerators can be laid on the bottom adjacent to \•111
iiii
docks and main channels. Discussion with the
U.S. Army Corps of EngineersPhiladelphia
District has made it clear that navigational
interestswould object even to botto.mdiffusers •L• Switchgear
in anchorage
areasandmainchannels
o.fthe Fig.8. Miami Conservancy, District surface
Delaware. Moreover such installations would aerators.

interfere with maintenance dredging. Fortu-

nately it appearsthat even in port areasshort
stub diffusermanifoldsextendingnot more than bubble diffusershave often suffered from clog-
25 feet channelwardof the pier head line would ging, not only becauseof sedimentbut also
probably be acceptable.Installation of diffusers because of chemical deposits and bacterial
at the end of finger piers, whosesidesare used growths. The diffusersused in the Delaware
for mooring, appears to be especiallyadvan- tests, witl• their 5/32-inch orifices and ball
tageous (Figure 12). valves to limit entry o,fsedimentinto the mani-
As indicated previouslyit appears that dif- fold when not in use, are estimated to be op-
fusers could obtain much greater transfer ef- erable in rivers during critical seasonsfor 4
ficiency by using fine bubbles. However, this years prior to removal and cleaning.There is
insufficient information to conclude that this
possibilitystill remainsunverifiedfrom a prac-
tical viewpoint.In waste treatment plants fine relatively inefficient coarse bubble diffuser
representsthe optimum design,but it has been
used as a basis for budget estimatespending
further explorationof fine bubble diffuserpos-
sibilities.

SYSTEM COSTS

Somemajor studieshavepreviouslybeencon-
b ducted on the Delaware Estuary, and a great
deal of information has been made available
throughcooperatingagencies.
A systemanalysis
of BOD and DO conducted for this project
gave resultssimilar to those of the other in-
vestigators;i.e., to increasethe general
Fig. 7. Ruhrverband surface aerator. (a) mum level of DO from 3.0 to 4.0 mg/1 through-
Elevation. (b) Plan. A, aerator impellers; F, out the 40-mile critical area would require the
floats; S, drive shafts. additionof 350,000poundsof additionaloxygen
 Artificial Aeration o• Rivers 575

,. Channel
•. • --__

To
shore

Pile cluster

shore/ Submarine cable

Fig. 9. Single (a) and twin (b) aerator facilities.

daily. A system of aerators to accomplishthis ½/lb. If the mean cost for a complete system
taskwouldrequireat least50 aerationsites were 5.0 ½/lb, the total annualcostof adding
of various sizes,dependingon the differing re- the last rag/1of DO wouldbe about$2,400,000,
quirements of the various sections.For reasons rather than the $9,300,000indicatedby present
already outlined both mechanical and diffuser estimates based on effluent treatment alone.
aerators would be required. Cost estimatesare These costs would be materially reduced if
based on an equipment life of 15 years, a oxygenwere addedat a lower level of concen-
socialinterest rate of 6%, and an averageop- tration, say, at a minimum level of 2.0--3.0
erating period of 6 months, 3 months at half mg/1.
time (12 hr/day) and 3 months at full time
OXYGEN •)IFFUSION POSSIBILITIES
(24 hr/day). For surfaceaerators,oxygencosts
are indicatedto be 4.0 4/lb for three-unit sites, Possibilitiesof raising the DO level by dif-
4.6 4/lb for two-unit sites,and 6.6 4/lb for one- fusionof pure oxygenhave been reportedboth
unit sites.For diffuseraeratorsthe pier end unit for rivers [ChemicalWeek, 1969; Amberg et al.,
illustrated would furnish additional dissolved 1969']and, in muchmore detail, for wastetreat-
oxygen at 4.4 4/lb in 30 feet of water, other ment plants [Union Carbide Corporation,
types of diffuser installations costing 5.5-9.3 1970]. Although such possibilitieshave not
576 WHIPPI•E Al•D •

been fully explored,particularly with respectto bulence created by aerators, which causes
cost, studiesunderway indicate that for water dispersioneffectsmuch greater than those oc-
30-40 feet deep, oxygendiffusionmay provide curring in the natural stream. If this excess
an economical alternative to air diffusion under turbulence picked up into suspensionheavy
the conditionsassumed.However, further work recent depositsof fine clay, silt, and organic
would be required to determineoptimal design sludge,the long-term advantage of removing
aspects since there are several alternative such bottom deposits would very likely be
methodsof oxygen•ation.
Alsoit wouldbe neces- overcomeby the possibleshort-run harm done
sary to make an entirely different dispersion to biosystems.However, the critical section of
analysis since the dispersion analysis made the Delaware River, like that of most other
for the aeratorswould not be applicable. major rivers, is heavily used for navigation.The
aeration systemsconsideredwould be lesspow-
Ei•VIRONMENTAL ASPECTS
erful and create less disturbancethan the pro-
Induced oxygenationof rivers acts only to pellers of tugs and other craft using the river
eliminate DO deficiencyand must be planned in large numbers.The area involved is mainly
as part of a comprehensivepollution control 30 feet deep or more, over sandy bottoms or
program to be effective. A natural ecosystem sediment deposits long since consolidatedand
can be completelydestroyedor badly unbal- resistantto. erosionby turbulenceof this magni-
anced by chemical residues,heat, blankets of tude. It doesnot appear that there will be sub-
sludge, or excessivenutrients regardlessof the stantial adverse secondary environmental ef-
DO level of the water. The primary effectsof fects, but there will be large, immediate, and
river aeration and oxidation systems can be favorableprimary effects.
very favorable both to biosystemsand to, rec-
IN STITUTIONAL ASPECTS
reational values,provided that they are used
as a supplementto a general program of waste The German Ruhrverband was established
treatment. by federal legislationprior to World War I to
The secondaryenvironmentaleffectsof river provide a satisfactorysystemof water quality
aeration are less apparent. As discussedabove, control for the Ruhr valley as a whole. Mu-
either aerators or oxygenatorscan be operated nicipal and industrial interests that pay the
without any substantialdanger of direct local- expensesof the program are representedon the
ized damage to fish life. The main environ- Ruhrverband governing board. Accordingly
mental changeis the high degreeof river tur- when the governing board decided after a

Fig. 10. Surface aerator •nooring.

 Artificial Aeration o[ Rivers 577

r •x.• r thortries, and major industries also play a

considerablerole. In general, responsibilities
/•, , ,• • for planning and enforcementare diffuserather
than concentrated and rely heavily on public
opinion and political pressuret,o make effective
progress.These arrangementshave been fairly
successful,althoughnowherenear the anticipa-
ß II 11 tions of the public, but they are simply not
adapted to any kind of water quality program
except effluent lreatment. Use of various alter-
natives (including not only instream aeration
but also redistribution of effluents by trunk
sewers, treatment of storm street runoff and
'^ ,., s I• •1 s , • '• other 'unrecorded'BOD [Whipple, 1970], and
the grouping of small inefficient treatment
plants into collective seweragesystems as is
now the adopted policy in the State of New
Jersey) is arbitrarily precluded. Alternative
programs are not a matter of planning only;
Fig. 11. Movable surface aerator. A, aerator suitably enlightened individuals at the state
impellers; F, floats; P, propellers; D, diesel en-
gines; S, shafts; P, pilot house; T, fuel tank. or federal level, who do exist, can, if given the
means, prepare plans consideringsuch alterna-
tives.But how can suchplansbe implemented ?
critical drought in 1959 that installation of We now have complexarrangementsby which
aerators would be an economicalsupplement the costs of treatment programs are divided
to planned waste treatment programs,the de- among those concerned.If instream aeration
cisionwas implementedwithout delay. were adoptedin any case,it would reducethe
In England the fairly recently constituted costsof treatment.There are no arrangements,
river authoritiesprovide a singlewater quality however, by which the costs of the aeration
control authority in each major watershed would be borne by those spared the extra ex-
[Association o• River Authorities,1968; Craine, pensesof treatment. In fact, in most casesthe
1969]. Although some of the planning is re- planningof water quality enforcementis pro-
served for central agencies,these fiver authori- ceedingwithout sufficientanalysisto determine
ties with the concurrenceof the centralgo.vern- the relationshipsbetweeninducedoxygenetlon
ment could evaluate objectively the relative and various degreesof treatment. We seem to
meritsof a river aerationproposal,as comparecl be institutionallylocked into a system that
to those of an alternative plan relying on en- is very inflexibleregardingthe considerationof
tirely conventionalmeans, and could adopt alternatives,at a time when somepromising
whichever plan seemedpreferable. The fiver alternativesare being developed.
autho.ritiesappear to be exercisingtheir new There are, of course,a few U.S. agencies
powers conservativelyand cautiously,but at that have the power to conduct a basin-wide
least there is no institutional barrier to their water quality program. Conspicuousamong
adoptio.nof basin-widewater quality plans on these is the Delaware River Basin Commission.
their merits.
Althoughits actionsare limitedin practiceby
In the United States the situation is quite the four statesand onefederaldepartmentthat
different. Our institutionsare complexlygeared provide its governingbody membership., the
to only onebasicmethod,namely,effluenttreat- commissio.n
could,in theoryat least,adoptany
ment. A rather awkward procedureallowsa few comprehensivewater quality plan that met
exceptionsfor low flow augmentationby specific federal standards. The much less well-known
congressional•action. In the typical U.S. basin Miami ConservancyDistrict also has the re-
controlsand budgetaw responsibilityare di- sponsibility
for implementing
a water quality
ß

vided amongfederal, state, and municipalau- controlprogramfor a basin,subjectin this case

578 W/-IIPPLE AND ¾U

Pier

Fig. 12. Diffuser aeratorat pier end. (a) Elevation. (b) Plan.

to state water quality standards.The Miami ing the validity of our hastily adoptedwater
Conservancy District has determined that it quality standardsand the extent to which re-
will not be possibleto meet state standardsof quiredDO levelsshouldbe enforced.Whenever
water quality by means of secondary waste the degree of enforcementcontemplatedwill
treatment alonebecauseof the great population requiretreatment in excess, of normal secondary
and industrial growth in the basin. In addition standards,instream aeration will probably be
advancedwaste treatment, lo,wfio.w augmenta- found less expensive,often by large margins.
tion or induced oxygenationwill be required, Other technologicalalternativesshouldalso be
and the last alternative was estimated to be considered. Institutions should be developed
the least expensiveby far. Accordinglyin the through which such alternatives.can be con-
fall of 19'70a group of four aeratorswas placed sidered and adopted where warranted; other-
in the river and commencedoperation. These wise massive resource misalloqationsmay re-
aeratorswere financedby effluentchargesto be sult.
paid by municipalities
and industriesthat would Acknowledgments. Important contributions to
otherwise have had to resort to advanced waste
this work were made by Drs. J. V. Hunter and
treatment. F. W. Dirtman of Rutgers University, Dr. G. E.
Of course,questionsmay be raised concern- Mattingly of Princeton University, and Mr.
 Artificial Aeration o• Rivers 579

F. P. Coughlan, Jr., of Hazen and Sawyer, Con- Federal Water Pollution Control Administration,
sultants,New York City. Delaware Estuary comprehensive study, Pre-
Most of the analysis on which this article was liminary report and findings, Philadelphia,
based was supported by the Environmental Pro- Pennsylvania, 1966.
tection Agency, Water Quality Office, through Garland,
C. F., Research
ontheinfiuenc•e
of basin
demonstration project 16080 DUP, 'Oxygen Re- volume and geometry on performance of sur-
generation of Polluted Rivers,' and by the De- face-entrainment aerators, paper presented at
partment of Conservation and Economic Develop- 1969 American Society of Mechanical Engi-
ment, State of New Jersey. neers/American Institute of Chemical Engineers
Stream Pollution Abatement Conference, Rut-
REFERENCES gers University, New Brunswick, New Jersey,
1969.
Amberg, H. R., I). W. Wise, and T. R. Aspitarte,
Aeration of streams with air and molecular Hunter, J. V., and W. Whipple, Jr., Evaluating
instream aerators of polluted rivers, J. Water
oxygen, Tappi, 52(10), 1866-1871,1969.
Pollut. Contr. Fed., •2(8), 249-262, 1970.
Association of River Authorities, Yearbook, 15
Imhoff, K. R., Oxygen management and artificial
Great College Street, Westminster, London,
reaeration in the area of Baldeney Lake and
SWl, England, 1968. the lower Ruhr River, Gas Wasser•ach,Beil.,
Barnhart, E. L., Transfer of oxygen in aqueous 109, 936, 1968.
solutions,J. Sanit. l•'ng. Div., Amer. $oc. Civil Kneese, A. V., and B. T. Bower, Managing Water
Eng., 95(SA3), 645-661, 1969.
Quality: Economics, Technology, Institutions,
Bewtra, J. K., and W. R. Nicholas, Oxygenation
Johns Hopkins Press, Baltimore, Maryland,
from diffused air in aeration tanks, J. Water 1968.
Pollut. Contr. Fed., 36(10), 1195-1224,1964.
Tarassov, V. J., H. J. Perlis, and B. Davidson,
Carver, C. E., Oxygen transfer from falling water
Optimization o,f a classof river aeration prob-
droplets, J. Sanit. Eng. Div., Amer. Soc. Civil lems by the use of multivariable distributed
Eng., 95(SA2), 239-251, 1969.
parameter control theory, Water Resour. Res.,
Chemical Week, For rivers breathing room, 131- 5(3), 563-573, 1969.
132, June 1969.
Union Carbide Corporation,Investigationof the
Craine, L. E., Water managementinnovationsin use of high purity oxygen aeration in the con-
England, Resourcesfor the Future, Washing-
ton, D.C., 1969.
ventional activated sludge process,186 pp.,
WaterPollut. Contr.Res.Set.,17050DNW05/70,
Davidson, B., and R. W. Bradshaw, A steady Federal Water Pollution Control Administra-
state optimal design of artificial induced aera-
tion, Washington,D.C., 1970.
tion in polluted streams by the use of Pontry- Whipple, W., Jr., BOD mass balance and water
agin's minimum principle, Water Resour. Res.,
quality standards, Water Resour. Res., 6(3),
.6(2), 383-397, 1970.
827-837, 1970.
Davis, R. K., The range o.f choice in water man- Whipple, W., Jr., et al., Instream aeration of
agement, a study of dissolved oxygen on the polluted rivers, Water Resources Research In-
Potomac estuary, Johns Hopkins Press, Balti-
stitute, New Brunswick,New Jersey, 1969 (ob-
more, Maryland, 1968.
tainable as PB 192637from Clearinghousefor
Doffman, D., and J. Westman, Responses of
Federal Scientific and Technical Information,
some anadromous fishes to varied oxygen con- Springfield,Virginia).
centrations and increased temperatures, 2,
Final Completion Rep. Proj. B-O12-N.J., Water
Whipple, W., Jr., F. P. Coughlan,and S. L. Yu,
Instream aerators for polluted rivers, J. Sanit.
ResourcesResearch Institute, Rutgers Univer-
Eng. Div., Amer. Soc. Civil Eng., 96(SA5),
sity, New Brunswick, New Jersey, 1970. 1153-1165, 1970.
Eckenfelder, W. W., Absorption of oxygen from Wright, J. F., and R. Porges,Water quality plan-
air bubbles in water, J. Sanit. Eng. Div., Amer. ning and management experiencesof the Dela-
$oc. Civil Eng., 85(SA4), 89-99, 1959. ware River Basin Commission, Proc. 5th Int.
Eckenfelder, W. W., and D. L. Ford, New con-
Water Pollut. Res. Conf., pp. I-3/1-I-3/17,
cepts in oxygen transfer and aeration, in Ad- 1970.
vances in Water Quality Improvement, edited
by E. F. Gloyna and W. W. Eckenfelder, 513 (ManuscriptreceivedJanuary 11, 1971;
pp., University of Texas Press, Austin, 1968. revised February 16, 1971.)